Instructions to use XHToken/Spark-X2.5-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use XHToken/Spark-X2.5-4B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="XHToken/Spark-X2.5-4B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("XHToken/Spark-X2.5-4B", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use XHToken/Spark-X2.5-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "XHToken/Spark-X2.5-4B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "XHToken/Spark-X2.5-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/XHToken/Spark-X2.5-4B

SGLang

How to use XHToken/Spark-X2.5-4B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "XHToken/Spark-X2.5-4B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "XHToken/Spark-X2.5-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "XHToken/Spark-X2.5-4B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "XHToken/Spark-X2.5-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use XHToken/Spark-X2.5-4B with Docker Model Runner:
```
docker model run hf.co/XHToken/Spark-X2.5-4B
```

Does the model support multiplingual?

by Duonglv - opened 4 days ago

Discussion

Duonglv

4 days ago

Hello XHToken,

This model seems to be a very strong model under 12B. Extremely excited. But does it support multilingual languages, such as Vietnamese?

Duonglv

4 days ago

And please quantize it to FP8 or NVFP4 for maximum size reduction.

linchao-2026

SparkLLM org 4 days ago

Hi @Duonglv , thanks for the kind words ,

On language support: Yes — Vietnamese (vie) is fully supported. The model card currently only tags English and Chinese, but the model actually covers 200+ languages. The table below shows some of the languages we've focused our optimization on:

Code	Language
`vie`	Vietnamese
`cmn`	Mandarin Chinese
`eng`	English
`tha`	Thai
`ind`	Indonesian
`zsm`	Malay
`fil`	Filipino
`jav`	Javanese
`khm`	Khmer
`lao`	Lao
`mya`	Burmese
`tam`	Tamil
`tet` / `tdt`	Tetum / Tetun Dili
`jpn`	Japanese
`kor`	Korean
`por`	Portuguese
`fra`	French
`deu`	German
`rus`	Russian
`spa`	Spanish
`arb`	Standard Arabic

On quantization: Thanks for the suggestion — FP8 releases are on our radar. We'll release them in the next few days！

Duonglv

4 days ago

Great. I think the community really needs truly small models like yours. The best size may be in the range of 1B to 36B (to fit in 32G VRAM).

And please add a multilingual tag/description to your model card. It would be very helpful and would even make the model look stronger.

Thanks a lot!

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment